研究目的：失眠盛行率高，且容易發展為慢性化的問題，目前多以助眠藥物作為失眠的治療，短期使用雖可幫助緩解失眠症狀，但長期使用的有效性及安全仍有較大爭議，且容易發展出依賴性及耐受性的問題。然而失眠患者中仍有相當高的比例在對藥物療效不滿意的情況下，使用遠超過臨床指引建議的期間，許多失眠患者亦表達出對助眠藥物使用的矛盾心態，顯示助眠藥物的使用仍受其他心理因素影響。過去研究指出計劃行為理論能有效預測及幫助探討物質使用相關行為的認知因素，但助眠藥物的使用除了認知因素的影響外，也會出現較為衝動及難以控制的渴想狀態。因此本研究目的是在於了解長期助眠藥物使用的心理因素，探討助眠藥物使用行為的認知因素，以及在計劃行為理論的架構下，失眠患者對於助眠藥物使用的認知因素是否會受到渴想的影響。 研究方法與結果：本研究以21名失眠患者進行開放式訪談後編制助眠藥物使用之計劃行為理論問卷，招募失眠且有助眠藥物使用經驗的受試者填寫助眠藥物使用計劃行為理論問卷以及助眠藥物渴想量表，並於施測後一個月及三個月的時間點，以電話追蹤其助眠藥物使用狀況，一個月及三個月有效問卷分別為143及139份。結果顯示知覺行為控制為行為意圖的主要預測因子，行為意圖及知覺行為控制能預測一個月及三個月時的藥物使用行為，且行為意圖為使用行為的最佳預測因子；助眠藥物使用行為意圖對一個月及三個月時的助眠藥物使用情況的預測力會受到渴想的負向調節。 結論：本研究結果顯示助眠藥物使用的行為意圖是行為出現與否的主要決定因素，說明失眠患者的確會在認知決策後選擇使用助眠藥物，也會在知覺到較多促進助眠藥物使用之情境或因素時產生助眠藥物使用的行為意圖或直接出現助眠藥物的使用行為，顯示助眠藥物使用行為上外在情境及因素的重要性。此外，本研究結果也顯示助眠藥物使用的行為意圖會受到渴想的干擾，當渴想程度愈高時行為意圖對於實際行為的預測力愈低，而渴想程度愈低時行為意圖對實際行為的預測愈高。此研究結果幫助了解長期助眠藥物使用的心理因素，亦可將其用於幫協助擬定助眠藥物減藥策略。 / Insomnia has high prevalence and is prone to develop into a chronic problem. Hypnotics are the most common choice of treatment for insomnia. Although short-term usage is effective in relieving insomnia, the safety and efficacy of long-term usage are still in controversy. Long-term usage has also been reported to increase risks of tolerance and dependence. Even though the efficacy of hypnotics is not always satisfying, the duration of hypnotics use in insomnia patients is often longer than clinical recommendation. This may be influenced by some psychological factors. The theory of planned behavior has been demonstrated to be able to predict cognitive factors of substance use behaviors. It can therefore possibly be used to model hypnotic use behavior. In addition to cognitive factors, craving that is associated with losing control over drug use may also play a role in the use of hypnotics. The purpose of current study is to 1) explore the cognitive factors of long-term hypnotics use based on the theory of planned behavior, and 2) understand whether the cognitive factors would be influenced by the craving state. Method and result: We first conducted a semi-structured interview with 21 insomnia patients to develop the questionnaire of the theory of planned behavior of hypnotics use and a hypnotic craving scale. The insomnia patients who had used or are currently using hypnotics were recruited to complete the questionnaires. They were contacted by telephone at 1-month and 3-month follow-ups. Total number of valid questionnaires of 1-month and 3-month follow-ups were 143 and 139, respectively. We found that behavior intention could be predicted by perceived behavior control, but not attitude and subjective norms. Frequency of hypnotics use at 1-month and 3-month follow-ups could be predicted by behavior intention and perceived behavior control, and behavior intention was the best predictor. Craving has a negative moderating effect on the relationship between behavior intention and the hypnotics use behavior at both 1-month and 3-month follow-ups. Conclusion: The findings of current study show that behavior intention is the most crucial cognitive factor to predict hypnotics use behavior. It indicates that insomnia patients may have higher behavior intention and are more likely to use hypnotics when they perceive more situational factors that lead to drug use. In addition, our study indicates the association between behavior intention and actual hypnotic use behavior can be moderated by craving. The higher the craving state the lower the predictability of hypnotic use by behavior intention, and vice versa. Our study helps understand the psychological factors of long-term hypnotics use, and assists in developing the tapering interventions.